SSP Ensemble Review

Seasonal Patterns in Sea Surface Temperatures

Author

Adam Kemberling

Published

July 27, 2023

Reviewing the Bias-Corrected Ensemble Datatsets from CMIP6

This report exists to visually review the bias-corrected CMIP6 data ensembles. Ensembles are from two Shared Socio-economic Pathways (SSPs): 1. SSP 1 (Sustainability) 2.6
2. SSP 5 (Fossil Fueled Development) 8.5

This document will look into three things. 1. How do these two pathways compare for our study areas 2. How do they compare against real-life observations 3. Are there regional biases or errors

Additionally, I also plan to flag points in time when each SSP ensemble passes temperature thresholds for the different areas. Bias corrected timelines for the sea surface temperature, sea surface salinity, bottom temperature, and bottom salinity will be compared against the reference observations (OISSTv2, SODA) to check how accurately they represent observed conditions, and whether that changes between our areas of interest.

Focus Areas

For clarity on what data is included/attributed to each region, the areas we focused on have been plotted below.

Load the Data

Bias-Corrected SSP Ensemble Data

Both the bias-corrections and the observational datasets (OISSTv2, SODA) have been prepared separately and are loaded below.

CMIP6 Bias-Corrected Arrays

Bias Corrected Ensembles - Reference Period (2000-2019)

The following figures chart out how each variable changes in time across the different models within an area. This data has been locally bias-corrected using observational datasets (OISSTv2) and reanalysis models (SODA).

Historic Observation Datasets

There are two data sources with which the CMIP6 model outputs are checked against to provide bias-corrections. These are the OISSTv2 and SODA reanalysis datasets. Monthly-averaged timeseries for each region can be viewed below.

Comparing Scenarios to Observations

Enhanced Biological Data

To fit the VAST model a suite of covariates matching the biological observations are uploaded. These values should fall within the ranges of the observed Temp/Salinity values and SSP ranges. If they don’t

In the following figures the points are the annual average of the point-extracted seasonal covariates.

VAST covariate scaling and prediction un-scaling:

Rescaled Enhanced Biological Data and Projected Physical Data – Before we fit the model, we center and scale each variable.

This dataset is at > Box/Mills Lab/Projects/COCA19_Projections/data/combined/all_tows_all_covs_rescale.rds.”

We keep the mean/SD info for each variable and then also go through a process to rescale the projected values from the CMIP6 experiments using these mean and SD values before making the projections.

These “prediction” datasets are located in > Box/Mills Lab/Projects/COCA19_Projections/data/predict”

They are named: 1. CMIP6_SSP5_85_mean_dat.csv 2. CMIP6_SSP5_85_95thpercentile_dat.csv 3. CMIP6_SSP5_85_5thpercentile_dat.csv 4. CMIP6_SSP1_26_mean_dat.csv 5. CMIP6_SSP1_26_95thpercentile_dat.csv 6. CMIP6_SSP1_26_5thpercentile_dat.csv

@Adam Kemberling do you think we could make similar ribbon plots from these datasets as the ones you’ve produced already? The y axis will be different because it is now the scaled/centered seasonal average BT and SST, but we could double check that general patterns and trends should match what we see with Adam’s current ribbon plots. If for some reason we don’t, that points to a potential issue with the CMIP6_data_processing.R script, or potentially the “scaling/centering” approach.

How do the Scaled Seasonal Values look?

Scenario Distances from Baseline Observations: 2000-2019

A 20-year period was used as a “baseline period” for fitting species distribution models to environmental features of surface and bottom temperatures in space.

In the following tables the average annual temperatures for this period have been removed, with temperatures reported as a departure from this baseline.

These tables could be slimmed down a lot. I think the most interesting things to include would be the means from both scenarios, and the distance between the two of them: “SSP1: 2.6 projects a mean temperature of +1C by 2040 in the Gulf of Maine. The more pessimistic scenario SSP5: 8.5 projects a similar future at +1.2C”… Something like that. Range of confidence and amount they overlap would also be interesting

Ensemble Differences from Observed Baseline Conditions
Difference in Bias-Corrected Ensemble Data when Compared to 2000-2019 Averages from OISSTv2 & SODA
SSP1: 2.6 SSP5: 8.5
Ensemble Mean 5th Percentile 95th Percentile Ensemble Mean 5th Percentile 95th Percentile
Canadian Survey Area - Surface Temperature
2000 −0.50 1.48 1.48 −0.50 1.48 1.48
2010 −0.11 1.57 1.57 −0.11 1.67 1.67
2020 0.29 2.13 2.13 0.37 2.46 2.46
2030 0.57 2.52 2.52 0.91 3.19 3.19
2040 0.75 2.81 2.81 1.51 4.02 4.02
2050 0.95 3.17 3.17 1.86 4.55 4.55
2060 1.00 3.33 3.33 2.52 5.36 5.36
2070 0.85 3.39 3.39 3.19 6.63 6.63
2080 0.75 3.28 3.28 3.69 7.18 7.18
2090 0.83 3.39 3.39 4.37 8.05 8.05
2100 0.89 3.46 3.46 4.77 8.60 8.60
EPU_SS - Surface Temperature
2000 −0.55 1.40 1.40 −0.55 1.40 1.40
2010 −0.15 1.56 1.56 −0.15 1.63 1.63
2020 0.29 2.12 2.12 0.37 2.42 2.42
2030 0.56 2.45 2.45 0.90 3.12 3.12
2040 0.79 2.76 2.76 1.54 3.88 3.88
2050 0.98 3.17 3.17 1.89 4.43 4.43
2060 1.04 3.27 3.27 2.54 5.15 5.15
2070 0.87 3.30 3.30 3.23 6.37 6.37
2080 0.74 3.25 3.25 3.72 6.88 6.88
2090 0.80 3.32 3.32 4.39 7.74 7.74
2100 0.88 3.22 3.22 4.80 8.20 8.20
EPU_GOM - Surface Temperature
2000 −0.50 1.17 1.17 −0.50 1.17 1.17
2010 −0.16 1.40 1.40 −0.11 1.49 1.49
2020 0.32 1.95 1.95 0.41 2.26 2.26
2030 0.57 2.29 2.29 0.91 2.94 2.94
2040 0.77 2.62 2.62 1.54 3.71 3.71
2050 0.97 2.96 2.96 1.97 4.36 4.36
2060 1.04 3.03 3.03 2.63 5.21 5.21
2070 0.93 3.16 3.16 3.34 6.45 6.45
2080 0.79 3.09 3.09 3.87 7.00 7.00
2090 0.88 3.19 3.19 4.59 7.92 7.92
2100 0.98 2.99 2.99 4.99 8.60 8.60
EPU_GB - Surface Temperature
2000 −0.36 1.44 1.44 −0.36 1.44 1.44
2010 −0.01 1.63 1.63 0.02 1.73 1.73
2020 0.29 2.09 2.09 0.53 2.41 2.41
2030 0.55 2.46 2.46 1.00 3.07 3.07
2040 0.78 2.72 2.72 1.55 3.83 3.83
2050 0.90 3.07 3.07 1.84 4.28 4.28
2060 0.91 3.14 3.14 2.43 5.00 5.00
2070 0.77 3.07 3.07 3.09 6.14 6.14
2080 0.66 2.98 2.98 3.51 6.61 6.61
2090 0.61 2.95 2.95 4.15 7.47 7.47
2100 0.71 2.92 2.92 4.47 8.01 8.01
EPU_MAB - Surface Temperature
2000 −0.33 1.46 1.46 −0.33 1.46 1.46
2010 −0.06 1.57 1.57 0.00 1.78 1.78
2020 0.20 1.95 1.95 0.48 2.40 2.40
2030 0.53 2.29 2.29 0.89 2.96 2.96
2040 0.68 2.48 2.48 1.34 3.60 3.60
2050 0.75 2.81 2.81 1.73 4.06 4.06
2060 0.76 2.77 2.77 2.31 4.69 4.69
2070 0.67 2.77 2.77 2.93 5.55 5.55
2080 0.60 2.68 2.68 3.33 5.98 5.98
2090 0.58 2.60 2.60 3.94 6.73 6.73
2100 0.81 2.64 2.64 4.10 6.88 6.88
Ensemble Differences from Observed Baseline Conditions
Difference in Bias-Corrected Ensemble Data when Compared to 2000-2019 Averages from OISSTv2 & SODA
SSP1: 2.6 SSP5: 8.5
Ensemble Mean 5th Percentile 95th Percentile Ensemble Mean 5th Percentile 95th Percentile
Canadian Survey Area - Bottom Temperature
2000 −0.28 1.09 1.09 −0.28 1.09 1.09
2010 0.13 1.36 1.36 −0.01 1.29 1.29
2020 0.57 2.08 2.08 0.48 2.17 2.17
2030 1.00 2.56 2.56 0.97 2.72 2.72
2040 1.37 3.05 3.05 1.63 3.62 3.62
2050 1.75 3.80 3.80 2.06 4.26 4.26
2060 1.96 4.09 4.09 2.77 5.09 5.09
2070 1.97 4.34 4.34 3.42 5.96 5.96
2080 1.97 4.42 4.42 3.95 6.55 6.55
2090 1.97 4.44 4.44 4.69 7.54 7.54
2100 2.09 4.57 4.57 5.07 8.16 8.16
EPU_SS - Bottom Temperature
2000 −0.26 0.91 0.91 −0.26 0.91 0.91
2010 0.06 1.14 1.14 −0.05 1.12 1.12
2020 0.44 1.73 1.73 0.41 1.87 1.87
2030 0.88 2.29 2.29 0.83 2.37 2.37
2040 1.12 2.58 2.58 1.48 3.13 3.13
2050 1.47 3.39 3.39 1.90 3.76 3.76
2060 1.60 3.44 3.44 2.57 4.43 4.43
2070 1.65 3.65 3.65 3.20 5.52 5.52
2080 1.58 3.73 3.73 3.90 6.23 6.23
2090 1.60 3.80 3.80 4.69 7.29 7.29
2100 1.68 3.69 3.69 5.16 7.90 7.90
EPU_GOM - Bottom Temperature
2000 −0.29 0.97 0.97 −0.29 0.97 0.97
2010 0.04 1.20 1.20 −0.08 1.26 1.26
2020 0.46 1.85 1.85 0.29 2.00 2.00
2030 0.78 2.21 2.21 0.71 2.47 2.47
2040 0.94 2.47 2.47 1.34 3.21 3.21
2050 1.32 3.26 3.26 1.76 3.82 3.82
2060 1.40 3.27 3.27 2.40 4.56 4.56
2070 1.45 3.51 3.51 3.07 5.75 5.75
2080 1.38 3.50 3.50 3.78 6.44 6.44
2090 1.40 3.49 3.49 4.54 7.21 7.21
2100 1.52 3.36 3.36 5.02 7.80 7.80
EPU_GB - Bottom Temperature
2000 −0.13 1.16 1.16 −0.13 1.16 1.16
2010 0.19 1.50 1.50 0.05 1.45 1.45
2020 0.58 2.16 2.16 0.46 2.15 2.15
2030 0.93 2.52 2.52 0.92 2.75 2.75
2040 1.24 2.99 2.99 1.48 3.57 3.57
2050 1.52 3.56 3.56 1.88 4.09 4.09
2060 1.70 3.78 3.78 2.49 4.85 4.85
2070 1.72 3.97 3.97 3.14 5.89 5.89
2080 1.66 4.02 4.02 3.69 6.49 6.49
2090 1.68 4.09 4.09 4.45 7.49 7.49
2100 1.76 4.22 4.22 4.69 7.99 7.99
EPU_MAB - Bottom Temperature
2000 −0.03 1.59 1.59 −0.03 1.59 1.59
2010 0.32 2.01 2.01 0.21 1.99 1.99
2020 0.71 2.76 2.76 0.65 2.67 2.67
2030 1.14 3.16 3.16 1.13 3.24 3.24
2040 1.35 3.60 3.60 1.58 3.98 3.98
2050 1.60 4.11 4.11 2.04 4.49 4.49
2060 1.68 4.28 4.28 2.68 5.25 5.25
2070 1.73 4.43 4.43 3.34 6.17 6.17
2080 1.63 4.44 4.44 3.80 6.73 6.73
2090 1.70 4.48 4.48 4.50 7.67 7.67
2100 1.80 4.61 4.61 4.69 8.11 8.11
Ensemble Differences from Observed Baseline Conditions
Difference in Bias-Corrected Ensemble Data when Compared to 2000-2019 Averages from OISSTv2 & SODA
SSP1: 2.6 SSP5: 8.5
Ensemble Mean 5th Percentile 95th Percentile Ensemble Mean 5th Percentile 95th Percentile
Canadian Survey Area - Surface Salinity
2000 0.02 0.86 0.86 0.02 0.86 0.86
2010 0.00 0.75 0.75 0.00 0.79 0.79
2020 −0.04 0.76 0.76 −0.02 0.81 0.81
2030 −0.09 0.69 0.69 −0.09 0.81 0.81
2040 −0.16 0.79 0.79 −0.13 0.87 0.87
2050 −0.25 0.78 0.78 −0.35 0.65 0.65
2060 −0.36 0.64 0.64 −0.39 0.67 0.67
2070 −0.40 0.62 0.62 −0.53 0.71 0.71
2080 −0.42 0.65 0.65 −0.75 0.59 0.59
2090 −0.50 0.62 0.62 −0.91 0.52 0.52
2100 −0.40 0.92 0.92 −1.00 0.54 0.54
EPU_SS - Surface Salinity
2000 −0.01 0.92 0.92 −0.01 0.92 0.92
2010 −0.02 0.85 0.85 −0.01 0.88 0.88
2020 0.00 0.91 0.91 0.02 1.02 1.02
2030 −0.05 0.83 0.83 −0.04 1.02 1.02
2040 −0.08 1.00 1.00 −0.05 1.09 1.09
2050 −0.14 1.02 1.02 −0.28 0.82 0.82
2060 −0.28 0.85 0.85 −0.27 0.86 0.86
2070 −0.31 0.84 0.84 −0.37 1.01 1.01
2080 −0.32 0.82 0.82 −0.61 0.86 0.86
2090 −0.43 0.88 0.88 −0.77 0.78 0.78
2100 −0.34 1.20 1.20 −0.87 0.86 0.86
EPU_GOM - Surface Salinity
2000 −0.05 0.82 0.82 −0.05 0.82 0.82
2010 −0.07 0.75 0.75 −0.07 0.77 0.77
2020 −0.05 0.78 0.78 −0.05 0.86 0.86
2030 −0.12 0.73 0.73 −0.10 0.86 0.86
2040 −0.14 0.89 0.89 −0.11 0.96 0.96
2050 −0.20 0.89 0.89 −0.31 0.66 0.66
2060 −0.33 0.75 0.75 −0.29 0.73 0.73
2070 −0.34 0.77 0.77 −0.38 0.86 0.86
2080 −0.36 0.68 0.68 −0.61 0.72 0.72
2090 −0.44 0.77 0.77 −0.76 0.65 0.65
2100 −0.34 1.09 1.09 −0.82 0.74 0.74
EPU_GB - Surface Salinity
2000 −0.14 0.67 0.67 −0.14 0.67 0.67
2010 −0.09 0.66 0.66 −0.08 0.72 0.72
2020 −0.08 0.76 0.76 0.02 0.91 0.91
2030 −0.08 0.77 0.77 0.00 0.96 0.96
2040 −0.08 0.91 0.91 0.02 1.08 1.08
2050 −0.17 0.93 0.93 −0.20 0.79 0.79
2060 −0.27 0.78 0.78 −0.18 0.84 0.84
2070 −0.30 0.73 0.73 −0.22 0.98 0.98
2080 −0.33 0.71 0.71 −0.41 0.87 0.87
2090 −0.47 0.71 0.71 −0.50 0.87 0.87
2100 −0.39 0.99 0.99 −0.56 0.92 0.92
EPU_MAB - Surface Salinity
2000 −0.19 0.45 0.45 −0.19 0.45 0.45
2010 −0.14 0.43 0.43 −0.13 0.49 0.49
2020 −0.12 0.48 0.48 0.02 0.73 0.73
2030 −0.05 0.59 0.59 0.03 0.75 0.75
2040 −0.01 0.72 0.72 0.06 0.92 0.92
2050 −0.07 0.73 0.73 0.02 0.81 0.81
2060 −0.09 0.70 0.70 0.09 0.94 0.94
2070 −0.10 0.71 0.71 0.15 1.10 1.10
2080 −0.12 0.70 0.70 0.10 1.10 1.10
2090 −0.21 0.74 0.74 0.08 1.18 1.18
2100 −0.13 0.86 0.86 0.05 1.40 1.40
Ensemble Differences from Observed Baseline Conditions
Difference in Bias-Corrected Ensemble Data when Compared to 2000-2019 Averages from OISSTv2 & SODA
SSP1: 2.6 SSP5: 8.5
Ensemble Mean 5th Percentile 95th Percentile Ensemble Mean 5th Percentile 95th Percentile
Canadian Survey Area - Bottom Salinity
2000 −0.04 0.31 0.31 −0.04 0.31 0.31
2010 −0.01 0.29 0.29 −0.01 0.30 0.30
2020 0.00 0.34 0.34 0.04 0.42 0.42
2030 0.06 0.41 0.41 0.09 0.47 0.47
2040 0.11 0.51 0.51 0.15 0.59 0.59
2050 0.14 0.59 0.59 0.18 0.67 0.67
2060 0.13 0.60 0.60 0.28 0.77 0.77
2070 0.15 0.67 0.67 0.33 0.89 0.89
2080 0.15 0.68 0.68 0.35 0.95 0.95
2090 0.14 0.70 0.70 0.39 1.07 1.07
2100 0.18 0.77 0.77 0.45 1.25 1.25
EPU_SS - Bottom Salinity
2000 −0.05 0.32 0.32 −0.05 0.32 0.32
2010 −0.05 0.26 0.26 −0.04 0.27 0.27
2020 −0.04 0.31 0.31 0.03 0.41 0.41
2030 0.01 0.38 0.38 0.05 0.42 0.42
2040 0.07 0.49 0.49 0.13 0.57 0.57
2050 0.08 0.56 0.56 0.14 0.63 0.63
2060 0.05 0.51 0.51 0.22 0.69 0.69
2070 0.06 0.54 0.54 0.27 0.80 0.80
2080 0.05 0.54 0.54 0.31 0.88 0.88
2090 0.04 0.60 0.60 0.34 0.96 0.96
2100 0.05 0.64 0.64 0.37 1.13 1.13
EPU_GOM - Bottom Salinity
2000 −0.03 0.46 0.46 −0.03 0.46 0.46
2010 −0.03 0.35 0.35 −0.02 0.37 0.37
2020 −0.02 0.40 0.40 0.03 0.51 0.51
2030 0.02 0.47 0.47 0.05 0.52 0.52
2040 0.06 0.62 0.62 0.11 0.68 0.68
2050 0.06 0.67 0.67 0.09 0.67 0.67
2060 0.00 0.57 0.57 0.15 0.71 0.71
2070 0.00 0.57 0.57 0.19 0.84 0.84
2080 −0.02 0.54 0.54 0.19 0.86 0.86
2090 −0.04 0.63 0.63 0.18 0.87 0.87
2100 −0.05 0.67 0.67 0.19 1.04 1.04
EPU_GB - Bottom Salinity
2000 −0.05 0.38 0.38 −0.05 0.38 0.38
2010 −0.03 0.31 0.31 −0.02 0.34 0.34
2020 0.01 0.42 0.42 0.05 0.49 0.49
2030 0.05 0.45 0.45 0.09 0.53 0.53
2040 0.13 0.62 0.62 0.16 0.70 0.70
2050 0.12 0.66 0.66 0.19 0.72 0.72
2060 0.11 0.63 0.63 0.27 0.77 0.77
2070 0.11 0.67 0.67 0.32 0.93 0.93
2080 0.12 0.69 0.69 0.34 0.99 0.99
2090 0.10 0.73 0.73 0.37 1.13 1.13
2100 0.16 0.85 0.85 0.41 1.28 1.28
EPU_MAB - Bottom Salinity
2000 −0.09 0.37 0.37 −0.09 0.37 0.37
2010 −0.04 0.35 0.35 −0.03 0.38 0.38
2020 −0.02 0.42 0.42 0.05 0.53 0.53
2030 0.07 0.49 0.49 0.11 0.58 0.58
2040 0.15 0.66 0.66 0.16 0.75 0.75
2050 0.14 0.71 0.71 0.23 0.80 0.80
2060 0.14 0.73 0.73 0.33 0.93 0.93
2070 0.14 0.78 0.78 0.42 1.10 1.10
2080 0.15 0.82 0.82 0.46 1.24 1.24
2090 0.15 0.89 0.89 0.52 1.40 1.40
2100 0.20 0.99 0.99 0.54 1.60 1.60

Seasonal Comparisons

Do baseline conditions look vastly different on a map?

Spring = March - May Summer = June - August August = September - November

3 seasons x 2 variables x 3 (observation dataset + 2 scenarios)